Content

This is challenge is about testing time series forecasting algorithms (Lectures 4, 5, 6, 7, 8 and 10).

The purpose of this challenge is to test your ability to use many models to forecast different time series and to evaluate them.

The analysis you perform at this first stage will be the basis to perform accurate forecasts during the next challenges.

Requirements

You are required to produce a fully reproducible notebook report explaining all the steps to forecast with Time Series, Machine Learning and Ensemble models.

You are free to apply the feature sets you created during challenge 1.

Data

Challenge dataset contains:
- daily website page views
- hourly website organic searches
- weekly website sales